Han Wang

Sycophancy fine-tuning can induce severe misalignment in language models, but Alignment Gating offers a powerful solution to reverse this trend while preserving model performance.

DPA4 achieves quantum-mechanical accuracy with a fraction of the parameters and training cost, setting a new standard for interatomic potentials.

Current speech generation models still fall short in maintaining consistency and capturing nuanced expressiveness when generating long-form speech, despite advances in high-fidelity synthesis.

LLMs can generate GPU kernels, but they're surprisingly bad at it: 72% of fusion tasks fail across all methods, and nearly half of the "correct" kernels are actually slower than PyTorch.

Kohn-Sham eigenvalues, often dismissed as unphysical, actually represent quasiparticle bands after accounting for dynamical core excitations, resolving a decades-old discrepancy with ARPES measurements.